论文部分内容阅读
以GE-E3高压涡轮第一级气冷导叶为研究对象,通过多目标遗传算法优化气膜孔布局以降低叶片表面温度。其中采用源项法模拟全场气膜冷却效果,该方法节省计算量的同时无需对气膜孔划分网格,通过对比实验结果后认为,源项法可以较好地模拟出冷气覆盖效果如表面动量损失等。在此基础上采用多目标遗传算法NSGA-II(Nondominated Sorting Genetic Algorithm II),以气膜孔的出气角及流向位置为设计变量,以叶片表面最高温度及平均温度为优化目标。结果表明叶片表面温度分布有所改善,其中压力面优化效果要好于吸力面。
Taking the first stage gas-cooled guide vane of GE-E3 high-pressure turbine as research object, the gas-hole layout was optimized by multi-objective genetic algorithm to reduce the blade surface temperature. The source term method is used to simulate the entire field film cooling effect. The method saves the calculation amount and does not need to divide the film holes. After comparing the experimental results, the source term method can well simulate the effect of the cooling air cover, such as the surface Momentum loss and so on. Based on this, a multi-objective genetic algorithm (NSGA-II) is used to design the outlet angle and the flow direction of the gas film holes as the design variables, and the maximum temperature and the average temperature of the blade surface are optimized. The results show that the temperature distribution of the blade surface has been improved, of which the pressure surface optimization effect is better than the suction surface.